1. Field of the Invention
The present invention relates generally to providing power to power-consuming systems and, more particularly, to a user-configurable power architecture with hot-pluggable power modules.
2. Related Art
Many businesses are highly dependent upon the availability of client/server network systems to permit essential network services and functions to be carried out. Because such network services have become increasingly essential to everyday business operations, steps need to be taken to ensure continuous availability of the network systems. A primary concern in achieving this objective is to minimize or eliminate server downtime.
A common cause of server downtime is temporary or permanent loss of line power. Utility power failures and fluctuations can decrease server availability due to replacement of damaged hardware and/or reinstallation of software applications. Inconsistency in power availability also causes system lockups, crashes, lost data and faulty data transmissions. Consequently, state-dependent computer systems such as servers are inherently vulnerable to damage and data loss due to interruptions in electrical power.
To prevent data loss and the attendant inconvenience of server state loss, traditional servers have been supplemented with uninterruptible power supplies (UPSs). There are internal UPSs configured, for example, in the form of a plug-in card for installation within the server chassis. There are also external UPSs which are connectable in series with the server line power. A UPS typically contains a battery, a recharge circuit and a detection and switching circuit. The recharge circuit operates to maintain the battery in a charged state. The detection and switching circuit operates to provide battery power to the server upon detection of an interruption in line power. This maintains server availability at least for a limited time during which steps can be taken to safeguard information on the server from eventual loss when the UPS battery is exhausted.
There are a number of disadvantages associated with the use of external UPSs. For example, external UPSs require the use of additional desktop or floor space, along with additional power cables to connect the UPSs to the servers. This is particularly problematic in data centers in which many hundreds of servers can be collocated. In particular, rack-mounted systems are particularly difficult to support with external UPS devices due the number of servers that can be vertically arranged in a relatively small area of the data center. In addition, to provide the server with information regarding the status of its power source, external UPSs often have serial communications capabilities, which are connected through an external serial port on the server. Since many computer systems are configured with only one or two external serial ports, this communications link between the server and the UPS consumes a limited resource.
There are drawbacks associated with existing internal UPSs as well. For example, the batteries of the internal UPS must be accessible for exchange when the batteries exceed their operational life or otherwise need to be maintained. This requires power to server to be interrupted to enable the operator to open the server chassis and remove the UPS card. Such interruption to the server power causes an interruption in server availability.